Non-ferrous Extractive Metallurgy

Prof. H. S. Ray
Department of Metallurgical and Materials Engineering
Indian Institute of Technology, Kharagpur

Module No. # 01
Lecture No. # 10
Refining of Metals - Chemical Methods

So far, we have been discussing principles of methods of production of metals using

Pyrometallurgy, Hydrometallurgy and Electrometallurgy. We will now discuss some
techniques of refining, in other words, we would like to go from impure metal to pure metal.

(Refer Slide Time: 00:46)

Now, I think, we should first be very clear as to what we mean by the word purity.
Theoretically speaking, there is nothing in this world which is totally pure; it is because we
have no method of knowing if something is totally pure. You know that according to
Avogadro’s hypothesis, 1 gram atom of anything will have 6 into 10 to the power 23 atoms,
but our purity detection level, say in the chemical laboratory, at the most can be finding 1 in
10 to the power 9, that is 1 parts per billion.
Now, that implies that 6 into 10 to the power 23 minus 9, that is 6 into 10 to the power 14. So
many atoms of impurity in a gram atom will go undetected. So if there are other atoms ,many
impurity atoms that will not be detected. We cannot detect impurities below 1 part per billion.

So, if we find something and we find it absolutely pure according to today’s analytical
technique then the impurity level is at best below 1 part per billion but not zero. We can never
say that a metal is totally pure; there are other impurity atoms below that level, which we are
not able to detect. Fortunately, perhaps often impurities at such levels may not be injurious.
As a matter of fact we should say that an impurity is an impurity when it has an undesirable
effect on the intended use of the metal. Consider for example copper which is a metal often
used in electrical applications.

(Refer Slide Time: 03:49)

See, if copper is for electrical applications then the oxygen is a very serious impurity, it has to
be less than 0.01 percent sometimes. Why? Because, if you allow oxygen to be there, then the
electrical conductivity is very badly affected, some oxides precipitate at the grain boundaries.
Hence oxygen needs to be avoided. If, however, one wants to make a sculpture out of
copper alloy or copper, say, copper in coinage, it does not matter if there is some oxygen
because; it does not affect the intended use. So, we always look at purity, from the point of
view of the question as to whether the impurity is affecting the intended use, if it is not
affecting the intended use, we can ignore that. But then, in many situations we find that we
have impurity elements, which must be remove or, otherwise, we cannot use the metal for the
intended use.

Now, the first thing to know is that, there are two basic approaches in refining of impure or
crude metal. In the first approach one tries to first produce bulk metal that is impure and then
attempts purification. The aim is to produce as large a quantity as possible first. After the bulk
recovery, one worries about the impurity contents. In the second approach, right from the
beginning we try to produce a pure compound and which will automatically give pure metal.
So, you see the two basic approaches are quite different from each other.

What do we do in iron and steel making? We do not worry about producing pure iron, pure
steel or pure anything. We just produce as much of iron as possible as pig iron, it has a whole
lot of impurities-- carbon, manganese, silicon, phosphorous, Sulphur etc.. We then go for
steel making which a purified version of pig iron. So, we are first producing bulk quantities,
then we are purifying Elsewhere we will find examples where we cannot use this approach.
Right from the beginning, the aim would be not to produce metal in the bulk but a pure
compound and once you have got the pure compound from there automatically the metal you
will get will be pure.

(Refer Slide Time: 07:59)

There are, many ways of doing it, and we can say that purification can be achieved by
chemical methods or by physical methods. I will very quickly indicate some of these methods
and what I am going to say is not a total picture, but some of the important ones I would
certainly mention.

Under chemical methods, the most important method is fire refining--- a method that has
been used for thousands of years. Fire has been used to eliminate, from the crude metal,
impurity elements or elements we do not want. Consider iron and steel, simply by keeping it
under fire in the molten state and by injecting oxygen through that we can remove carbon,
silicon, phosphorous sand manganese from iron. This is how pig iron is purified in iron and
steel industry to produce steel. Steel is a purified version of pig iron; we eliminate carbon,
silicon, phosphorous, manganese as stable oxides by blowing oxygen in the open hearth
furnace or basic oxygen converter.

The idea is to produce oxides of carbon, silicon, phosphorous, manganese, which form
preferentially prior to forming of iron oxides. Of course in the process, the pool of metal gets
very saturated with dissolved oxygen which has to be removed later on by a process of DE
oxidation. That is a separate matter.

Now, similarly, we can remove a whole lot of impurities in other case e.g.--, lead, arsenic,
antimony, tin, Sulphur, magnesium, aluminum, iron, etcetera from copper. One treats molten
copper by an oxidizing atmosphere, and the impurity elemental form either volatile oxides or
solid oxides which separate out as sludge. We can remove sodium from zinc, antimony,
arsenic, zinc from tin, and zinc from lead. We can also remove iron, aluminum, zinc from tin,
and zinc from gold.

Now, in all these cases, the things that are being removed, obviously, are forming more stable
oxides compared to the metal that you want to purify. When we come to discussing
production and purification of individual metals, some of these will be repeated.
(Refer Slide Time: 11:53)

A very effective way of removing impurities is through chlorination. We can remove zinc
from lead, because if you have lead and zinc in the molten state and if you pass chlorine, zinc
chloride will form vapours that come out. lead chloride will not form when zinc chloride is
forming. Zinc, copper, lead can be removed from bismuth, lead from tin and, silver, copper,
zinc from gold etcetera.

So, the basic principle in this sort of fire refining is very simple. We are keeping the crude
metal in the molten state under oxidizing conditions; we control the partial of pressure of
oxygen to various levels and remove the impurity elements either as volatile oxides or as
solids which would come out as a slag to be skimmed of periodically.
(Refer Slide Time: 13:52)

The other methods of that will make use of chemical processing, of course would be in
hydrometallurgy precipitation if everything is taken into solution it can precipitate, impurity
elements as oxides or carbonates or sulphates or whatever. Then, there are some special
processes call ion exchange and solvent extraction, exclusively used in Hydrometallurgy,
widely used in Hydrometallurgy, I will discuss that little later.

Under chemical methods of purification, one should also include electro refining, there the
idea is that, in a suitable electrolyte, we will have the impure metal as anode, we will have a
sheet of pure metal as cathode and we will do electro refining. When this impure metal M
will dissolve as M plus ions or whatever with the balancing, this will go and deposited on the
pure cathode, so which time this will begin to become thinner and this will become thicker.

So, from an impure thick anode, we will produce a pure thick cathode. What will happen to
the impurities that are there in the impure or crude metal? Two things can happen: from here,
some impurities may dissolve in the electrolyte, this is electrolyte, some impurities will
dissolve, some will precept, some will not dissolve, but as the metal is taken out of this, they
will simply drop as insolvable particles called sludge. So, two things can happen, we will
produce a pure cathodes from the impure anode, impurities will be released some will go into
the solution, some will be deposited at the bottom or sludge.
Now, both of them may be very valuable, we can recover by-products from the sludge; we
can recover some impurities, if they have their uses from the solution also, so these are some
of the chemical methods.

(Refer Slide Time: 17:32)

Now, before I go into a solvent extraction and electro refining, let me discuss a very
interesting method, which come under the class of vapour transport processes. Vapour
transport process, this is not suitable for large scale large scale production, but it is an
extremely sophisticated technique for producing pure metals from impure metals and it is
applicable for many metals.

But, the reason I am discussing is, it is based on a beautiful thermodynamic concept, see a
whole idea here is, that we will take an impure metal, make it react with a vapour or gas, say
we call it vapour or gas 1 to produce another vapour 2 means, that vapour which is the
gaseous reactant reacts with the impure metal and takes it into a gas phase, vapour 2 at some
temperature T1. This vapour is taken out and at another temperature, T 1 can be greater than
T 2 or sometimes T 1 can be less than T 2, this reaction is reversed means, we will produce,
we will generate the vapour 1, but now we will have pure metal, I hope you understand this
scheme.

We have impure metal and we create an intermediate gas phase vapour 2 using a gaseous
reactant vapour 1 and then this intermediate gas phase that we have created is at another
temperature, which may be higher or lower decompose to regenerate that vapour and produce
pure metal. Obviously, this thing cannot be done in the same chamber, the idea is that, once
you have produced the vapour 2, you take it out into another chamber and then you
decompose to get the pure metal and this vapour of course will be recirculated.

In the source system, there is no consumption of the reactant which is taking the metal into
the gas phase; there is a very interesting example, application of this in the case of
purification of nickel. Now, that has a special name called carbonyl process, where nickel,
impure solid nickel is made to react carbon monoxide which is gas at about 45 degrees, at
such low temperatures it can be converted to nickel carbonyl. This nickel carbonyl at a higher
temperatures 200 and 40 degree centigrade decomposes and the reaction goes this way, so
you regenerate carbon monoxide and now you have pure nickel; it looks very easy but, such a
process is very, very dangerous, carbonyl is a very poisonous gas everything has to be done in
a closed chamber.

Similar reaction is applied in the case of aluminum - liquid, impure one can react with
aluminum tetra chloride gas to form at 1000 degrees aluminum mono chloride gas, so using
this gaseous species, we are taking aluminum into a gas phase and then this gas decomposes
at a lower temperature to regenerate this and of course pure aluminum. So, essentially at one
temperature, we produce an intermediate gas by gasifying this metal to a reaction with a
gaseous species and then the reaction is reversed at another temperature, which can be at a
higher temperature or at a lower temperature.

(Refer Slide Time: 22:52)

Now a gentlemen called Van Arkel achieved some similar things around 1925 and it is called
Van Arkel’s process, very famous process, where impure tungsten solid is reacted with iodine
gas, I am writing I, because of the temperatures, where this reaction is taking place, iodine is
atomic to produce tungsten iodide gas at another temperature it will reversed.

This method can be applied for in a small scale for many metals like, beryllium, chromium,
copper, iron, nickel, palladium, silicon, thorium, uranium, vanadium, zirconium, etcetera, this
is the reason, why I want to discuss it in a few minute, because it is in a class of its own. A
very interesting process, some call it Van Arkel’s process, we have the other names like
carbonyl process for nickel, and in general, they are called vapour transport processes,
because we are transporting impure metal through a gas phase to a pure metal form; in this
cases, we are taking the crude metal at a lower temperature and we decompose it at a higher
temperature.

Questions that come to our mind is, when will the temperature for the forward reaction T 1
and temperature for the T 2, when will T 1 be more than T 2? When will T 2 be less than T 1?
When this is achievable and when this is not achievable? What are the thermodynamic
conditions for that? I would like to discuss that.

(Refer Slide Time: 25:41)

But before I proceed, let me give to show you this diagram, this diagram figure shows how
purification of tungsten is done by Van Arkel’s process. These are the two tungsten
electrodes; this is a tungsten filament and by passing current, this filament is kept at a high
temperature. The rest of the chamber is at a lower temperature, but the filament inside is at a
higher temperature, because a current is passing through. This heat generated here may heat
the chamber, if necessary it can be heated from outside also.

Now, iodine comes to this chamber separately and we have a vacuum arrangement to
maintain whatever pressure we want, what happens is that the crude metal is kept along the
periphery, in a cage we have molybdenum supporting grid, means there is a cage all around
but there are perforations, so the gases can come and go.

Now, in this case, we have seen the temperature required for forming tungsten iodide, I have
not given the temperature, but I will give it little later, it forms at a lower temperature. So, in
this chamber tungsten iodide vapour will form and when it goes and hits the filament, if this
region it gets heated, so pure tungsten is deposited and then iodine vapour is rejected; so the
iodine vapour will keep coming here, react with to form tungsten iodide and move there and
it will get decomposed tungsten when deposit.

So, you transport titanium tungsten from this cage to the filament, the filament grows thicker,
this filament grows thicker and from here, you lose crude metal gradually, whatever
impurities are there in the tungsten will be left behind in this place.

Now, this is one design, essentially we are able to have in the same chamber two temperature
zones: one zone is here, another zone is here and the way it is done is very convenient that the
crude metal stays here and the pure metal goes at another place and iodine is coming here
picking up a tungsten from here, going there and getting disassociated the tungsten iodide and
it keeps on going on and that material goes, when we have got a very thick filament, we will
cut it off put another tungsten wire and we will put more tungsten here a crude tungsten and
we will do the purification.
(Refer Slide Time: 29:04)

Of course, it is a a very schematic diagram, the rate of deposition will depend on many
things; rate of deposition, rate of deposition will depend on factors like geometry,
arrangement of filament with respect to crude feed. One has to optimize the distance between
the filament and the crude feed, the dimensions of the chamber, it would depend on a nature
of feed material, it will depend on temperature of the vessel, at what temperature, and it will
depend on the pressure of iodine gas that is maintained within the chamber etcetera, etcetera,
and etcetera.

(Refer Slide Time: 30:38)

Let us look at this reaction in little more detail. Essentially the reaction is that, a metal X solid
or liquid is reacting with a gas Y to generate an intermediate gas, this is the intermediate gas.
Now, there are y moles of gas giving only one mole of gas, in this side we have more entropy,
because the gas is disappearing in the reaction, creating less gas we have less entropy, so we
have an entropy change in this which is negative.

Now, we can write for the forward reaction, delta g naught 1, minus r t ln p x Y y we are
dealing with two gaseous species y y, it is happening at temperature T 1 delta H naught 1
minus T 1 delta s naught 1. And for the backward reaction, we can similarly write ln p x Y y p
Y y. Now, I have mentioned before that normally delta h naught, delta s naught values do not
change with temperature, so we can actually get rid of the notations of 1, 2, etcetera with
delta h naught minus T 1 delta s naught, delta h naught minus t delta s naught.

(Refer Slide Time: 33:50)

Now see, kind of things we can derive from the simple looking at the simple thing, what we
really want is that, a situation like this, we need to ensure that the free energy change in the
reaction or free energy change formation of the intermediate will go from negative to positive
side, so for the forward reactions, it should be below 0 in this direction, below this
temperature, beyond this temperature, this line must cross the 0, so that the reverse reaction
becomes possible.

You will recall that, in the case of Ellingham diagrams, where free energies of formations of
oxides plotted against temperature, basically it was metal always in terms of 1 or 2 and we
are writing 2 M O, the picture is somewhat similar. In this also entropy is reduced, because a
gas phase on the left hand side disappears, so there is an entropy change and that is why all
the lines move upwards and I had always mentioned that, this is the entropy term, same thing
here the slope here will depend on the entropy term.

So, basically, you go back to the previous picture, the reversal of the reaction whether we can
go from in the forward direction and then at another temperature, come to the left hand side,
it will depend on several factors. First of all, suppose there was no heat of reaction, if delta h
naught was 0 then T 1, T 2, T 2 T 2 cancel out, which means, the reaction becomes insensitive
to temperature changes.

If there was no entropy change, then it would not be possible to reverse the reaction at all
because, delta G naught does will not depend on delta on temperature anymore, temperature
would get cancelled out; on the other hand, if the delta h naught term was very large, then
also will be a problem, because delta h naught actually is the intercept also in Ellingham
diagrams you had seen.

It means, if the line was something like this, the temperature was so high, that it was
inconceivable that will go to such a high temperature for decomposition, so there is a
restriction here, should not be 0, should not be too high; interestingly this is a critical term
because larger the slope of this line, the faster will be able to reverse that, but if it is very
shallow, then something we have might have to go to very high temperature reverse. But, if it
is strip then, we should quickly be able to ensure reversal and that will depend on the entropy
change, which will depend on what kind of intermediate gaseous species you are making, you
are able to generate.
(Refer Slide Time: 37:51)

So, this is the equation and as I have mentioned, if there is no delta H naught in the reaction,
then T 1 would get cancelled out and therefore, this term will not depend on delta on
temperature anymore, reaction becomes insensitive. This is critical, this will tell us how fast
or how slow the things can happen (Refer Slide Time: 38:14).

(Refer Slide Time: 38:17)

Now look at the actual data, for reactions of tungsten chloride to form w Cl 6, this is for
tantalum to put TaCl5, this is hafnium to form hafnium tetra chloride. Now all these seeds can
be, impure metals can be purified by this method, form an intermediate gas then, disassociate
it with a higher temperature to get pure metal and regenerate the highlight. I am giving you
data for chlorine and this data are given for different pressures in the chamber for one
atmosphere lines 10 to the power minus 3 atmosphere lines.

Let us ignore that, let us look at the solid lines only, you will find the slope is more, slope of
the line is more for tungsten, because here 6 moles of gas are consumed to make one mole of
gas, this lot more entropy change, delta s naught is larger; in this case, 5 moles react with the
solid to produce one mole of gas, in this case, 4 moles make 1 mole of gas. So, you see the
changes in slope of these lines accordingly, it is highest here, it is intermediate here, and it is
it has the least slope here, so we will try to ensure that, we form intermediate gas that gives us
maximum change of entropy.

Now in this case, because tungsten has a higher valency, if the process becomes very
effective for tungsten, because in temperatures below 1000 degrees, we can produce this gas
just we go beyond 1000 degrees and the reaction is reversed, I hope you have understood the
basic principle.

(Refer Slide Time: 40:46)

I have given some data and you will find that same thing here, that the delta H naught value
for tungsten 2 W 6 is higher than that for tantalum pentachloride, which is large and that for
hafnium. So, the delta is not the critical parameter this also high is important, because when
this is small like in the case of tungsten hexa chloride the intercept is small, means here very
quickly going to you can quickly cross the 0 point in delta g naught scale.
The larger it is the further away from that line that you have to cross and of course, the these
are also important, the melting point parameters, because if tungsten for example, had the
melting point been low then what would happen, that in the filament you will not have the
solid depositing, you might get a liquid and the liquid will drop off and you would not get a
solid filament you will have all kinds of problems; the fact is that, it has a high melting point,
it definitely helps and the in the in the kind of apparatus we have. So, there are all this
parameters are important in order to show that the most important thing is the entropy
parameter.

(Refer Slide Time: 42:26)

Let me just show you another slide here, as I said we are bringing in iodine or chlorine gas,
but we are writing in terms of Cl because, the temperatures where we are forming the
intermediate gas and the Cl2 dissociates, so we can write the reaction as MCl giving you the
intermediate gas where n, n can have different values for different metals. For this we have an
expression like that, that delta g naught plus RT l n k and that R T l n k will have 3 terms: one
pressure of this intermediate gas, ln p this divided by PCl and of course the activity of m
which is one, we do not we can forget about that.

If the pressure in the chamber is 10 to the power minus 3, we finally get an expression like
this and when M p m cl n intermediate gas is also in the chamber 10 to the power minus 3,
this will be the final expression. So, this gives us the slope and entropy change in
decomposition reaction given by n becomes decisive factor as you can see will given by this
reaction.

Now the dissociation temperatures for various metals are like this, for beryllium 800 degrees,
chromium, copper 900, iron you can see, nickel, silicon, thorium, titanium, uranium,
zirconium, tungsten will decompose tungsten chloride 6 only at 1100 degrees, this is all for
10 to the power minus 3 atmosphere equation.

(Refer Slide Time: 44:29)

Because, if we change the pressure from this to that, then actually we are creating a, for us if
we look at the solid line this is for 10 to the power minus 3 no this sorry, this is the solid lines
are for one atmosphere and dashed lines are 10 to the power minus 3. For solid line, this is if
we have greater pressures, then we have a solid line and if we have a reduced pressure we
have the dashed line. Now, if you operate at a very low pressure, you see, you might have to
go to very high temperatures for decomposition, whereas we will be able to do that at a lower
temperature if you had higher pressures.

But there are many, many factors to be taken into account as to whether you should operate
with high pressures or low pressures; maybe we need not discuss it now. But, the this gives
you the idea that you can do the reversal reactions over a wide range of temperature. One has
to take into account kinetics of that process also with what will be better at high pressures,
what will be better to low pressures.
(Refer Slide Time: 46:12)

Now, there are some general rules for such transport processes and I have mentioned some of
them; first is that, no transport is possible if delta g naught value is 0 or very small. If the
enthalpy change for that reaction of formation of the intermediate gas phase is a very small,
then the process become insensitive. You cannot gasify the crude metal at one temperature
and decompose it another, because equilibrium constant becomes independent of T, again if
they absolute if they absolute value of delta g naught are very large, then there can be no
measurable transport, equilibrium constant becomes insensitive to temperature.

Like, if this is a very large value then, this term t 1delta naught term becomes to small
compared to this and we will not be able to have the reversal. The sign of delta h naught
determines the direction of vapour transport, whether it is endothermic, whether it is
exothermic will determine, whether you will form the vapour at a lower temperature and
decomposition a higher temperature or we will form it at a higher temperature and
decomposition at a lower temperature.

Like in the case of nickel, we form the intermediate gas carbonyl at a lower temperature
around 40 decompose it at 2 to 40, but in the case of aluminum, we go through a reverse
thing aluminum trichloride reacts with crude aluminum to produce aluminum mono chloride
at a higher temperature and it decomposes at a lower temperature to regenerate aluminum and
trichloride and a pure aluminum.
So you can see for yourself, how it is and by if you look at that equation, the sign will
determine the what will happen of course, for delta naught is equal not equal to 0, there is a
delta h naught for maximum transport. There has to be a balance between the values of delta s
naught, which is critically needed otherwise, we would not be able to have the reversal. There
is a value of delta h naught for maximum transport to take place and maximum transport
effect increases as the absolute value of delta s naught increases.

So, this is a brief discussion of the vapour transport processes, I discuss it in little more detail,
because thermodynamic principles are very interesting and this process makes use of this
thermodynamic principles, but it is not for a large scale purification of metals. Let me come
to two techniques called ion exchange and solvent extraction, who are extensively employed
in hydrometallurgy for purification of leach liquor.

In that case, we are not dealing with a solid impure metal and purifying it, the idea is to have
the metallic values in a solution along with a lot of impurity metals assigns. We would like to
selectively take out the metallic irons to produce a pure compound, from which we will
produce pure metal; we are coming to the second approach in purification.

(Refer Slide Time: 50:36)

Now, the ion exchange techniques makes use of some complex organic acids or bases, which
are insoluble in water the salts that are formed between, because of reaction of this resin and
a leach liquor, they are also insolvable in water.
So, basically I am showing you one particular cation exchange resin here, there are many
reason resins, these are all solid organic compounds and the structure has basic or acidic
groups, which are capable of ionization and exchange reactions; basically what you see here
is a large molecule that comprise of interconnected and cross length chains is a large
molecule, there is cross linking also and there is interlinking like this.

In this particular resin, this is the place where an exchange reaction will take place, this mean
hydrogen will be knocked off in that place a metal cation will be absorbed, so when you have
in the liquid many metals one or two particular metal, cation would be replaced by this,
hydrogen will be replaced by that and then, the metallic value that you are aiming at will be
observed by this resin. Now, here what we are showing is a cation exchange resin, because
this is where a cationic exchange will take place, we sometimes write is simply as r x means
resin matrix and x is the where the exchange will take place, r is the polymer backbone and x
is the ion exchange site.

Here the iron exchange site is S O 3 H, where the hydrogen will be replaced by the metal
cation by some reaction such as this, say copper sulphate solution is there, it comes in contact
with this resin, we are simply writing as R S O 3 8, because S O 3 this is the part where the
exchange reaction will take place.

(Refer Slide Time: 53:21)

So, in the place of h copper will go H 2 S O 4 will be created, so there are two step process,
the feed solution with metallic values will pass through this solid resin, unabsorbed impurity
ions will pass off unabsorbed, but the metallic value will be absorbed in the resin and then,
that can be washed or eluted through a process of washing or illusion by fresh water, fresh
acid, the resin will be regenerated and the solution will take out metallic values.

Now, this is the very simple explanation that basically what all we do, I will I will may be I
will I will come to that, when I continue with the lecture, let me now conclude this lecture.
We started by discussing purification techniques for metals, we said first of all, define what
do you mean by purity, unless the presence of an impurity is affecting the intended use, its not
impurity and then, there are different ways we can take out the impurity from a metal; one
approach is produced the metal in the bulk and then purify take out the impurity elements.

The other is do not follow that approach, to start with produce a pure compound and then
from that you get the metallurgy, there are whole lot of chemical methods. I have discussed
fire refining, sulphidation, chlorination, we have talked about intermediate vapour transport
process, then I have just started discussing iron exchange, which I will continue in the next
lecture, thank you very much.